Changes 1.8.3:
Fix build issue on Android
Remove the dummy driver
Add traces for MB rate control / temporal layer
Set verbosity level between {0, 1, 2} by setting the variable LIBVA_MESSAGING_LEVEL
in /etc/libva.conf or by setting the environment variable LIBVA_MESSAGING_LEVEL.
Changes that might impact existing code
Eigen 3.3 fixes a few shortcomings that might impact existing code:
* Eigen 3.3 clarifies the usage of the Index type in SparseMatrix,
PermutationMatrix, Transpositions, as detailed below.
* The normalize and normalized methods will now leave their input
unchanged if its norm is 0 or too close to 0. Previously this
resulted in a vector containing NaNs or infinities.
* internal::significant_decimals_impl is deprecated and users of
custom scalar types are encouraged to overload
NumTraits<>::digits10().
In Eigen 3.3, the evaluation mechanism of expressions has been
completely rewritten. c.f. http://eigen.tuxfamily.org/index.php?title=3.3
Looks as though C++11 support was added and Eigen 2 support removed.
Detailed changelog:
http://eigen.tuxfamily.org/index.php?title=ChangeLog#Eigen_3.3.4
Changes from 2.6.2 to 2.6.3
- Documentation now available at readthedocs.io_.
- Support for floor() and ceil() functions added by Caleb P. Burns.
- NumPy requirement increased from 1.6 to 1.7 due to changes in iterator
flags.
- Sphinx autodocs support added for documentation on readthedocs.org.
- Fixed a bug where complex constants would return an error, fixing
problems with `sympy` when using NumExpr as a backend.
- Fix for 277 whereby arrays of shape (1,...) would be reduced as
if they were full reduction. Behavoir now matches that of NumPy.
- String literals are automatically encoded into 'ascii' bytes for
convience.
Release 1.8.2:
Behavior changes related to build edges
* Detect build graph cycles as early as possible. Cycles containing single phony edges are now detected and warned about. The new flag `-w phonycycle=err` upgrades the warning to an error, while `-w phonycycle=warn` explicitly makes this a warning. Future versions of ninja will default to phonycycle=err. In a few years, we hope to make this an error unconditionally. CMake 2.8-3.0 produces ninja files that trigger this warning.,
* Always re-run edges that exit with non-zero exit code, even if they successfully write their output
* Support restat when rebuilding manifest
* Support mtime of 0 on input files
* Allow more path components
Windows-specific changes
* replace deprecated GetVersionEx with recommended replacement
* Add clparser_perftest
* Faster clparser, speeds up full builds by up to 20%
* Write subprocess output to stdout in binary mode
* Remove path component limit from input of CanonicalizePath in windows
POSIX-specific changes
* Use POSIX_SPAWN_USEVFORK with glibc, makes some full builds twice as fast
* Close original pipe fd in subprocesses
Changes to building ninja itself
* Fix build with uclibc
* Suppress warning 4244 to bootstrap on vs2017
* Add support for DragonFly BSD.
Changes to scripts
* Make zsh completion use explicitly specified ninja files
Bugfixes and minor changes
* Improve error message when a depfile contains a bad path
* fix a clang-cl -Wformat warning
* browse: Bind to localhost by default
* teach -t commands to optionally print only the final command
in ../libpreludedb via its Makefile.common. The extra one appears to
have been accidentally created while adding new distinfo hash
functions a couple years back.
This came to light because a new patch over the weekend didn't get
into this copy, and it seems that if a distinfo file appears in the
current directory, pkglint ignores the setting of ${DISTINFO_FILE}
even if it points somewhere else. Which is probably a bug.
Fixes in 1.65.1
Config:
* Version macro changes for CUDA.
* Update last known MSVC version.
* MSVC 14.11 supports structured bindings.
Context:
* Return a continuation from functions executed by resume_with.
Fiber:
* Return a continuation from functions executed by resume_with.
Smart Pointers:
* Fix Visual C++ version checks.
* Fix compilation for NVCC with host compiler clang.
Stacktrace:
* Update build testing file to detail with a MinGW issue.
* Change preprocessor file extensions to work with the installation system.
* Support FreeBSD and other OSs that do not define _GNU_SOURCE but have _Unwind_Backtrace function.
Thread:
* Fix bug in boost::condition_variable on Windows
Bug fixes only.
- Fixed bug in Session.merge() where an internal check for a target
object in the identity map could lead to an error.
- Fixed bug where an undefer_group() option would not be recognized.
- Fixed race condition in ORM identity map.
- Fixed bug in Session.merge() where objects in a collection that had
the primary key attribute set to None for a key that is typically
autoincrementing would be considered to be a database-persisted key
for part of the internal deduplication process.
- An InvalidRequestError is raised when a synonym() is used against an
attribute that is not against a MapperProperty.
- Altered the range specification for window functions to allow for
two of the same PRECEDING or FOLLOWING keywords in a range by
allowing for the left side of the range to be positive and for the
right to be negative.
2017-09-12 Richard Russon <rich@flatcap.org>
* Bug Fixes
- broken check on resend message
- crash in vfolder-from-query
* Build
- Be more formal about quoting in m4 macros
- fix warnings raised by gcc7
- notmuch: add support for the v5 API
pkgsrc changes:
- Delete 2nd hunk of patches/patch-audio_out_ao__oss.c, it was fixed uptsream
and it is no more needed for 0.27.0.
Changes:
0.27.0
======
Features
--------
Added
~~~~~
- Windows: handle media keys
- libmpv: options: add a thread-safe way to notify option updates
- vd_lavc/vo_opengl: support embedded ICC profiles
- vo: rendering API abstraction for future non-GL video outputs
- vo_opengl: add a gamut warning feature to highlight out-of-gamut colors (--gamut-warning)
- vo_opengl: add direct rendering support (--vd-lavc-dr)
- vo_opengl: implement (faster) compute shader based EWA kernel
- vo_opengl: implement HLG OOTF inverse
- vo_opengl: support HDR peak detection (--hdr-compute-peak)
- vo_opengl: support float input pixel formats
- vo_opengl: support loading custom user textures (#4586)
- vo_opengl: support user compute shaders
Removed
~~~~~~~
- Remove video equalizer handling from vo_direct3d, vo_sdl, vo_vaapi, and vo_xv (GPL, not worth the effort to support legacy VOs)
Options and Commands
--------------------
Added
~~~~~
- macOS: add --no-native-fs option to disable the native macOS fullscreen (#4014)
- player: add --track-auto-selection option
Changed
~~~~~~~
- input: use mnemonic names for mouse buttons, same as Qt: https://doc.qt.io/qt-5/qt.html#MouseButton-enum
- options: change --loop semantics
- player: make --lavfi-complex changeable at runtime
- vf_eq: remove this filter (GPL; uses libavfilter’s eq filter now, with changed semantics)
- video: change --deinterlace behavior
- vo_opengl: generalize HDR tone mapping to gamut mapping, --hdr-tone-mapping → --tone-mapping
Removed
~~~~~~~
- --field-dominance (GPL-only author, no chance of relicensing)
- input: drop deprecated "osd" command
- options: drop --video-aspect-method=hybrid (GPL-only)
Fixes and Minor Enhancements
----------------------------
- TOOLS/autocrop.lua: fix cropdetect black limit for 10-bit videos
- TOOLS/lua/autodeint: update to lavfi-bridge
- TOOLS/lua/status-line: improve and update
- af_lavrresample: don't call swr_set_compensation() unless necessary (#4716)
- ao_oss: fix period_size calculation (#4642)
- ao_rsound: allow setting the host
- audio: fix spdif mode
- filter_kernels: correct spline64 kernel
- macOS: fix media keys input when other Apps steal the priority (#4834)
- macOS: fix the support of multiple renderers (GPU switch) (#2371)
- macOS: remove the standard "Show Tab Bar" menu item
- options: fix --include (#4673)
- player: fix --end with large values (#4650)
- player: fix confusion in audio resync code (#4688)
- player: make refresh seeks slightly more robust (#4757)
- player: readd smi subtitle extension (#4626)
- vd_lavc: change auto-probe order to prefer cuda over vdpau-copy
- vd_lavc: fix device leak with copy-mode hwaccels (#4735)
- vd_lavc: fix hwdec compatibility with yuvj420p formats
- vd_lavc: fix mid-stream hwdec fallback
- vf_vapoursynth: fix inverted sign and restore 10 bit support (#4720)
- video: increase --monitorpixelaspect range
- vo_opengl: adjust the rules for linearization (#4631)
- vo_opengl: scale deband-grain to the signal range
- vo_opengl: tone map on the maximum signal component
- x11: fix that window could be resized when using embedding (#4784)
- ytdl_hook: resolve relative paths when joining segment urls (#4827)
- ytdl_hook: support fragments with relative paths, fixes segmented DASH
- The Nginx plugin now configures Nginx to use 2048-bit Diffie-Hellman
parameters.
- certbot-auto now installs Certbot in directories under /opt/eff.org.
- The Nginx plugin can now be selected in Certbot's interactive output.
- Output verbosity of renewal failures when running with --quiet has
been reduced.
- The default revocation reason shown in Certbot help output now is a
human readable string instead of a numerical code.
- Plugin selection is now included in normal terminal output.
- A newer version of ConfigArgParse is now installed when using
certbot-auto causing values set to false in a Certbot INI
configuration file to be handled intuitively.
- New naming conventions preventing certbot-auto from installing OS
dependencies on Fedora 26 have been resolved.
Maxima 5.40 change log
======================
New items in core:
------------------
[0eda309] New functions encode_time and decode_time, with test cases.
New items in share:
-------------------
[534a7d1] add symplectic_ode package
[46a7a0f] Add Kaplan-Meier's survival and empirical distribution estimators
Changes in core:
----------------
[cb259bb] change O(n^2) method for flattening list to O(n) method.
[21cf8e1] split totient into two parts
[6ef2b26] optimized gcfactor and tried to make it easier to read.
[1594167] optimize gc squaring and inline gctime1
[d7ff7c3] inline gctimes.
[4c8b6c6] sped up psumsq.
[f306940] speeded up imodp.
[efadb61] new implementation of ; is about a factor of 2 faster on sbcl
[9c79308] move jacbi and related functions from rat3c to numth
[75d8ceb] sped up divsum by splitting it in two parts where the lower
Numerous cross-compiling changes.
Numerous documentation changes and improvements.
Build/installation improvements.
Many updates to examples.
Many changes and improvements to test system.
Changes in share:
--------------
[9c7c70c] Commit new version of COMA. Sent by Wilhelm Haager
[0ef9ab0] Check for a constant "variable" of integration in romberg and bromberg
[834ffb3] Import ODEPACK fortran code.
[06bc1bd] Julia() and mandelbrot() are modified to create maxoutXXX.gnuplot
[dca80e3] plotdf: An option to change the arrow density. Patch from themusicgod1.
Bug fixes:
----------
[6ff1570] fix bug 3291: ecl on Linux needs utf-8
[db7788f] fix bug3300 Incorrect sha256sum output
[b896d1f] In rtest_translator, skip some tests which cause a crash (bug #3291).
[f1bc324] Implement numerical evaluation for gamma_greek. Fixes SF bug #3277.
[eec3d7f] In tests for kill(props), work around bug #3289.
[6267b88] In lsquares_mse, ensure temporary variable is assigned the correct value. #3282.
[b58a547] In lsquares_mse, ensure temporary variable is assigned the correct value. #3282.
[a7239b1] In RISCHLOGEPROG, detect zero divisor. This fixes SF bug #3274.
[eaacba6] Unlock :common-lisp package when compiling with SBCL. Fixes SF bug #3271.
[71ee1c0] Fix bug #2740: demo(romberg) fails
[5fb8f81] Add test for bug #2295: Failure to evaluate definite integral
[35c625e] Add tests for bug #2314: Two different results for an integral
[56fb70e] In tests for simplify_sum, add a few more tests from bug report #3236.
Unnumbered bugs:
----------------
[06c0fc7] operatingsystem: Make chdir work with recent gcl versions.
[7277f54] gcfactor(4/(%i+1)) gave an internal error in $bothcoef.
[c477689] Give informative error message when some index is not an integer: "rat(-1) isn't an integer?" from mailing list 2017-03-04.
[96c57d5] work around bug in gcl decode-universal-time.
[b4819b4] work around bug in gcl decode-universal-time.
[a36bf53] When evaluating float(diff(...)), apply float only to the first argument. Fixes bug reported to mailing list 2017-02-27.
[bffe54f] In GETL-LM-FCN-PROP, return NIL for argument which is not a symbol. Fixes bug reported to mailing list 2017-03-04.
[ef1fa9e] Remove built-in symbols from the list for user-defined properties. Fixes bug reported to mailing list 2017-02-24.
[70e05dd] Handle Lisp arrays in MQAPPLY expressions more carefully. Fixes bug reported to mailing list 2017-02-02.
[761cc1c] Handle Lisp arrays in MQAPPLY expressions more carefully. Fixes bug reported to mailing list 2017-02-02.
[b6dd08c] A solve() enhancement by Richard Fateman that should fix a zillion of bug tickets.
[b6b5c5c] Translate define and ":=" correctly by nuking the translator property. Fixes bug reported to mailing list circa 2017-01-24.
[5cfc6fd] Fix bernoulli and binomial pdfs when p=0 or p=1
[d6739e9] Fix bernoulli and binomial pdfs when p=0 or p=1
Maxima 5.39 change log
======================
New items in core:
------------------
New functions firstn and lastn
Changes in core:
----------------
Special variable clean-up
Crosscompiling support
Gnuplot 5.0 or newer versions
Documentation updates
function save: portable way to save "fast" arrays (hash tables)
functions timedate, parse_timedate: handle timezone offsets
Floating point overflow handling
Complex numericals
Testsuite additions and improvements
GIT fixes
ODE improvements
Load .wxm files from wxMaxima
Temporary file handling improvements
New in share:
-------------
package decfp: decimal bigfloat package (thanks to Richard Fateman)
Changes in share:
-----------------
In some share test scripts, put reset() at the top
package solver: rename solver.mac to Solver.mac
package stringproc: new functions slessp, slesspignore, sgreaterp, sgreaterpignore (string comparison)
package charsets: several updates and improvements; still work in progress
package boolsimp: code clean-up; no functional changes
package descriptive, functions continuous_freq, discrete_freq: faster counting algorithm; accept arrays as well as lists
Bug fixes:
----------
[54822ea] Fixed bug #3258: diff with negative order in integrate
[0e4ed41] Fix issue #3255 by changing how resimplification works
[55ba236] In risplit, in preparation of fixing issue #3255 (logarc)
[6e7f9dd] Partial fix#3216: In package lapack, load without compile
[ad86a80] Fixed bug #3232: In ezunits, don't compile predicate functions
[0ec66c6] Fixed GCL bug #49689: Mark a known failure in rtest11
[98b0349] Fixed bug #3213: Round when formatting bigfloats
[3745efa] Another test for bug #3244.
[67fd622] Fixed bug #3246: look-up entries for inv. trig antiderivatives
[120d8be] Fixed bug #3241: In MQAPPLY1, call AMPERCHK
[e916dd0] Fixed bug #3235: In RGRP, rephrase test for ECL
[7060221] Add tests from bug #3221
[2204fca] Fixed bug #3221: elliptic_pi(n,z,m) wrong for z > %pi/2
[cb3dc1c] Fixed bug #453: algsys fails in simple case
[7efa7eb] Add tests from SF bug 2059.
[4851e2c] Add tests from SF [bugs:2038] and [bugs:#2736]
640ca75 SF [bug:#1266][bug:#3208] Don't use $radcan when simplifying in algsys.
2986fc0 Fix SF [bugs:#3208]. Use sqrtdenest in algsys to simplify partial solutions.
5fd2f55 [bugs:#3208] algsys.lisp:presultant: resimplify result
4734ec8 share-subdirs.lisp is auto-generated *and* in version control - which is afaik unavoidable for the lisp-only build system. It also currently is out-of-sync with the repo => Updated the repo version.
f91d677 Increase output when running share/contrib/diffequations/tests
a5873ba SF bugs 3210 and 3212. Fix algsys regressions.
9a4694f Bug #2796 can't load ode2 with n declared constant
3c6c473 Bug report ID: 1621 Wrong solution to ode2
cf9cbb2 Fixed bug #2667: Distinguish trig function exponents in TeX
bf4cf94 In rtest_log, additional tests for SF bug #3105.
9ef3fee Fixed bug #3148: Most of the change is in indentation
8f523fc Fixed bug #3194: No simplification of tan(x+n*%pi) and cot(x+n*%pi)
fdadebc Fixed bug #3186 and other get_plot_option issues.
3fd2b4c Fixed bug #3022: Update the RPM spec file
aa0a365 Fixed bug #3180: declare(foo, antisymmetric)
9ad3f2f Add test for SF bug #3090
a57ff0c Add test for SF bug #3170
4568937 Fixed bug #3144: Avoid unbounded recursion in PTIMES%E
0bcccbe Fixed bug #3109: is(sin(x) <= 1) returns "unknown"
037ea4a Fixed bugs #3114, #3115: introduced by [647516] Fix integrate(x=0,x)
Unnumbered bugs:
----------------
[c8007b9] Partial fix to read_xpm.
[35790e2] Correct documentation for rest
[eb56614] Fix test for orderlessp transitive
[85aff4c] Fix in gcl-builds smismatch
[d37c3e4] Fix in timedate, round TZ
[74800c2] Remove redundant definition of copy from share/affine
[8ec3a05] Another fix for summation of rational functions
[394ca66] In package facexp, remove "&&" because
[b7ee537] Ensure that *OPR-TABLE* associates "{" with $SET
[9f43849] Ran update_examples on Simplification.texi
[dd6fbb1] Fix uses of complex-float-numerical-eval-p
[07363c2] Fix bug in running test files
[6d1b6fe] Partially fixed assumptions regarding products and powers
[639dd40] Partially fixed assumptions regarding products and powers
[691cbc4] Fix taylor testsuite failure bug introduced by [22f4fd]
Core Server
- Process responsible for running the autoheal partition handling
strategy could run into a deadlock with its peers, preventing autoheal
from completing.
- Garbage collection of mirrored queue metrics on nodes that did not
host a master or mirror for a queue affected delivery and
acknowledgement rates. This could result in rates being 0 or negative
when they should not be.
- Stats emission could prevent queue mirrors from performing garbage
collection and consume memory even when they were empty.
- RABBITMQ_SCHEDULER_BIND_TYPE and RABBITMQ_DISTRIBUTION_BUFFER_SIZE now
can be set via rabbitmq-env.conf.
Shovel Management Plugin
- Passwords in source and destination URIs are now redacted out.
Federation Management Plugin
- Passwords in upstream URIs are now redacted out.
